Retail Space Planning Community


Using Technology for Shelf Audits & Getting to the Root of Out-of-Stocks

By Mike Wilkening, Communications Manager, ARC

Shelf audit success — and getting to the bottom of out-of-stock and related issues — is key to a best-in-class space planning strategy. More and more, AI-powered technology is coming to the fore to help suppliers and retailers ensure the right products are in the right place at the right time.

Trax Retail is among the companies offering technological solutions to help a supplier’s best-laid planogram plans actually come to fruition. The Space Planning Community recently caught up with Matt Greene, Vice President of Enterprise Sales at Trax, to get a sense of how companies are leveraging tech at shelf. However, if you take one thing away from this Q&A, note how the importance of people shines through in succeeding in stores.

Editor’s note: This Q&A was edited slightly for clarity and brevity.

Space Planning Community: When CPGs/retailers come to you looking to get a better understanding of how to be more successful at shelf, what are the primary reasons?

Greene: It’s really two different main pieces. The first being, everyone is trying to be more efficient with their resources and their route to market. If they have a feet-on-the-street approach with a sales team and there are people in stores already doing store calls and selling in products and touching base with the retailers, they think, why not use that person that’s already there to gain really valuable market intelligence?

There are a lot of clients that use us with their own sales team as kind of a supplemental add-on to that store call that they’re doing that then really bifurcates into two different real use cases.

The first one is, how do you make that person that’s on that store call more efficient and gain a bunch of data in the process rather than that person walking in completely blind and assessing the state of what’s going on in the shelf with their competitors, what’s going on with their SKUs, what’s in the back room, etc.? They want to already have that information set up for them. So that basically improves the speed by which they’ll be able to do what they need to do in the store, the conversation that they need to have with the retailer or the manager of that store versus take the time to assess the universe and then react to it. They use our software to inventory the store, get the images and the data from the images served up to them prior to them walking into that store. Then that ultimately gives them a more efficient plan when they’re in that actual store.

The second piece is all around market intelligence. I used to work for Nielsen, and when you’re looking at the Nielsen point of sale scan data, there’s a lot of inferences that are drawn about what’s happening to drive that sales lift going up or going down. And many times, a CPG will kind of hide behind the story that they want to tell based off of, is it going up or is it going down?

There’s (also) been a higher amount of assortment entering the category from new and emerging brands that either found their space in the DTC (direct-to-consumer) world or had a slightly different economic model and got through COVID in a better way than maybe some of the other ones. There’s a lot of competition out there and there’s a lot of assortment in many of the categories that our clients use this in. Understanding the category through a shelf audit, or through image recognition from that shelf really allows them to understand the story of what’s happening in store beyond just the IRI or the Nielsen data telling them it went up or it went down.

Space Planning Community: What can image-recognition technology like Trax’s solve for – and how should it fit in a space planning approach for things like out-of-stocks and shelf audits?

Greene: I think one of the biggest misnomers of image recognition technology is that it is a holy grail for out of-stock fixes. The truth is, what it does is it shows you how bad the situation is, but it doesn’t quite give you any sort of root-cause understanding on why that is. Think of it like watching a train wreck live where you’re like, I see what’s going to happen here. This is not good, but there’s nothing I can do to stop this train. When you see how bad the shelves are through images and you can quite literally zoom in to holes in the shelf and see that your price tag is there but your product is not, it’s led to at least right now, this confirmation bias of, ‘Yeah, the shelves are really bad and we know that retailers are changing their policies around how much they can keep in the back.’

We know that theft has gone up significantly, as has this concept of phantom inventory, which basically means that all the ordering processes at the store level are really not dialed in. So I would say image recognition paints a picture of how bad the shelf is, but more often than not right now, it doesn’t actually paint a very different picture than you would’ve thought going into looking at the images.

Many of our clients say, ‘We’ve got a huge out-of-stock issue.’ And I would tell you that I don’t think the right answer is rush to image recognition to validate that because all you’re going to do is basically see what you already know now — you’re going to be able to get more detailed evidence, and you’ll be able to use those images of the shelf to present back to the retailer the main kind of cause of concern. But I don’t think that’s a holy grail to solving out of stocks and I’ll talk to you about how we are thinking about it out-of-stock resolution in a different way in a minute here. But my hunch is that the core two most important use cases for shelf auditing and image recognition, which I’ll kind of use synonymously, are around category optimization, seeing what’s happening in this totally changing tide of the category. And then, No. 2, sales rep efficiency, because you can get a lot of real value out of both of those. I don’t think you’re going to get as much value trying to fix the out-of-stock problem.

Space Planning Community: Can you give an example or two of how your tool helps retailers and CPGs communicate better with each other on issues like out-of-stocks?

Greene: We believe and we’ve seen that a picture tells a thousand words. So anytime you can extract a digital image and virtually walk stores through the tool that we have and basically show in real time as data comes in from the field what the realities of all these stores look like, that inherently is going to stoke a really strong conversation. Then it brings a real truth to the narrative, whatever that narrative is.

But we’ve found that especially right now, because supply chains are still a little out of whack, (and) backroom policies have changed a lot, and because basically the overall method of servicing retail has changed a lot too; you’ve probably heard of retailers (stripping) out almost all in-store labor and they rely specifically on third parties to basically keep things on shelf. And there’s a lot of retailers that are similar to that, maybe not to that extent. Most of the story actually lives within what’s going on in the back room to then present what the reality is on the shelf. How do you have a strong conversation with the retailer if you’re only going in with basically what’s on the shelf and photos and images and narratives and data points? That’s going to be a good conversation, but it’s really not going to solve the real core issues of how you keep a shelf looking great.

So what we’ve done is we’ve built a solution that we call signal-based merchandising. Trax has three main pillars to our company. The first one is all around shelf measurement, shelf and display measurement. That’s probably what you know us for and what we’ve kind of entered the market to be. The second is we have a merchandising capability where wehave an on-demand group of close to 40,000 field merchandisers that work on a 1099 contract model for us. Think of it like an Uber of merchandising. The hypothesis behind that is six, seven years ago, we kept having the same conversation come back to us from brands where they said the shelf looks terrible, I’m looking at it look terrible, it’s cool that you can tell me it looks terrible, but I need you to actually do something about this. So we built a merchandising capability operating in a consumer application to then go fix problems as we see them.

The third piece of what our company is, we have actually an engaged shopper network that uses an application, a consumer application called Shopkick. What that does is it basically incentivizes everyday consumers to buzz around the retail store, engage with certain SKUs, answers certain questions, and basically builds and heightens their level of understanding of a product that’s sitting on the shelf. We’re basically combining all three of those technologies now to get back to your point of what drives a strong conversation with a retailer.

Space Planning Community: What’s an example of a supplier-retailer conversation around out-of-stocks?

Greene: What we would do is if a brand perceived there to be huge out-of-stock issues and they wanted to as an example, walk up to (the retailer) and say, ‘Hey, we need to increase your inventory levels right now we don’t have enough product in the back, so therefore we’re going to have the product more front.’ That’s their hypothesis. What we would do is turn on the SKUs that that brand is concerned with in our shopper application. We would then start getting strong data points about where those SKUs are available and where they’re not available and where they’re not on the shelf. We would then send in our merchandisers to then go into those stores. The merchandisers would assess the shelf again, they would validate it’s not on the shelf, they’d go into the backroom, they would then assess what the backroom level looks like and then they would work with the store manager to look at their inventory on hand and put in orders to bring it to the store if it needs to be. That then builds a really strong data set for you to then say to the retailer, ‘Our product is nowhere on shelf in this 1,000 or 2000 stores.’ It’s not in the back room or if it is in the back room, it’s not on the front, which means you have an in-store labor (issue), which (is an) execution issue. And for that reason, my sales are going down with the end consumer causing my brand issues. So you have the same benefit of the category view, you have the same benefit of understanding what the shelf looks like, but you also then have backroom understanding, you have inventory on-hand understanding, and you start getting real trends that bubble themselves up at certain stores so you solve the root cause of the problem versus kind of just ‘attack the symptom’ as you see it. That’s when you then have a really good conversation with the retailer and ultimately what most of these conversations end with is the retailer saying, OK, you found the gap. We’re going to increase your inventory levels and just make sure you can service it once it hits the back room, make sure it gets on the shelf and that means that the velocity starts going up and you start seeing more and more sales and out-of-stock issues are less and less.

Space Planning Community: How closely do you work with retail space planners, floor planners in terms of their work and in terms of use of this tool?

Greene: That’s something we’re very, very keen and familiar with. I think because you can gain so many insights out of the data that we process, we find that a lot of our clients who are category captains who have real kind of sway in how the space looks in the store, they use our data and we have a product that kind of bifurcates out of our image recognition solution called Category Excellence, where we have predefined KPIs that are meant for that exact same use case. So things like blocking the brand, what specifically works, what doesn’t work, things like eye-level versus ankle-level – specifically, where do you want your brand to be to sell the most, what leads the aisle, what is the end of the aisle? There’s probably 50 KPIs that I could rattle off there, but what that space planner does is they basically blend in the deal center, the IRI sales data with the conditions that are on shelf, and the end goal, or the holy grail here, is what condition drives the most lift that doesn’t necessarily think about the execution side of it. I just talked about clearly to have lift, you need to have product on shelf. But assuming all product is sitting there, it becomes pretty clear to the space planner what works and what doesn’t work as they go into their category, kind of reset plan and figure out where it is they want to put the product, where they want to bring to the retailer, what works the best. They use our data to really validate a lot of that.

Space Planning: One last question. In your view, what does the rise of e-commerce mean for brands at shelf in a changing store/shopping environment?

Greene: The game has changed, but maybe not as much as people in just everyday life would think. So 85% of consumers still do want to shop in store (according to Trax data). So that 13, 14, 15% that are shopping online, obviously you need to pay special attention to what that looks like, but consumers still want to trust, they want to touch, they want to feel their product and I think there’s a bit of a narrative that it’s all gone. The truth is it’s not right. There’s a lot of people that still like being boots on the ground in the store now.

So I would even ask you, what’s the definition of a store anymore? To me it kind of feels almost more like a fulfillment center, like a warehouse, a really transactional kind of place vs. ‘Hey, go get my shopping cart and then go do what I need to do.’ The truth is people shopping in store too, it’s usually not the only store they’re going to be shopping in. I know myself personally, I’ll go to Target to hit certain things and then we’ll go to Whole Foods to do certain things and get produce and then we’ll have certain things delivered to us from another delivery model, whether it’s Target, whether it’s Amazon Prime. It’s taking on this weird transactional shape that has people diversifying a lot of what they do in the store.

I would say most important impact from that change is that a lot of the click-and-collect or pick-and-pack, whatever you want to call it about what happens in store, it actually comes from the center of the store. You’re not getting your orders fulfilled from the back room as much as you are. Someone is physically going in there to shop for you. If you’re using one of these models, they’re taking things off the shelf and that is impacting the everyday consumer because it’s product that you could have bought that maybe has gone through and because maybe the transaction model isn’t totally where it needs to be. That then creates the phantom inventory issue that I just talked about where the retailer might think, hey, the product’s there, but really it’s blown out because (delivery services) took it right from where the consumer would buy it.

If anything, I would say, point No. 1, people still want to shop in store. It’s still very important. Point No. 2, I think the consumer is diversifying how they shop. And then point No. 3, it’s creating a real execution gap for the people that are still in store. And that’s led to this really cool epiphany in the market of, we just talked about brands still want to know, how do you identify where the root cause of their issue is if the product isn’t on store? That kind of rinse and repeats everything we just discussed. You need to get the full story to keep the product there. It’s just taking it off the shelf through online e-commerce. It’s definitely impacting the customer experience.

Access our new Space Planning Team Benchmarking Survey and other exclusive members-only content in the Resource Library.

New to the community? Sign-up for our bi-monthly newsletter for more great space planning content.

Want to become a member to network with peers, train your team, share best practices, and more? Contact our member services team today.

About the Author: Mike Wilkening, Communications Manager, for the Association of Retail and Consumer Professionals (ARC).

Mike brings more than two decades of communications experience to the CMA/SIMA. He began his career in journalism, spending more than 10 years covering the National Football League for Pro Football Weekly and NBC’s Pro Football Talk. His bylines have also appeared in CBS MarketWatch, the St. Louis Post-Dispatch, NBC New York, and More recently, he has pivoted to corporate communications, including strategy and messaging experience for a Fortune 500 company. Mike holds a master’s degree in journalism from Northwestern University and a bachelor’s degree in accounting from the University of Illinois.